1. Field of the Invention
The present invention relates to a mass spectrometer for mass spectrometry (MS) based on ion flight distance in a given time being related to its mass-to-charge ratio. This has the advantages of time-of-flight mass spectrometry without the high-speed electronics normally required. The mass spectrometer may be in a tandem configuration to effect simultaneous collection of precursor and product spectra.
In its tandem mass spectrometer (MS/MS) configuration, the simultaneous production of the complete (MS/MS) spectrum for all the ions produced in the source provides an improvement in the efficiency and speed of mass spectrometric analysis as applied in biomedical research, drug delivery, environmental analysis and other applications.
2. Discussion of Related Art
Time-of-flight mass spectrometers are based on the difference in velocity attained by ions of different mass-to-charge ratios (m/z) when they are accelerated in a vacuum by an electric field. The common arrangement for the measurement of this velocity is to place a detector at the end of the flight path and determine the time required for the ion to reach the detector after acceleration. So, for a distance d between the acceleration region and the detector and a flight time of between the time of acceleration and detection t, the velocity v will be v=d/t. Since the distance is the same for all ions, their arrival times are different with the smaller m/z ions arriving first and the larger m/z ions later. This approach is called “time-of-flight” (TOF) mass spectrometry.
In the traditional linear TOF instrument, the ions would traverse a field free region at the end of which they would arrive at the detector in order of their m/z values. The detector signal intensity vs. time is recorded and presented as a mass spectrum.
Mass spectrometers can be devised to use either scanning mass-to-charge ratio (m/z) filters with a single detector (such as quadrupole or sector mass analyzers), batch m/z sorters with a single detector (such as the ion trap, the FTMS instrument or time-of-flight mass analyzers), or m/z spatial dispersion instruments with multiple detectors (such as a magnetic sector with linear detector array). When full spectral information is required, scanning filter instruments are the least efficient because they ignore huge portions of the sample ion beam while detecting the ions having the m/z value for which the filter is set at each instant.
Batch m/z sorting instruments are most efficient when the sample consumption is pulsed to coincide with the introduction of a new batch of sample ions. In cases where the sample comes in a continuous stream, as in chromatographic detection, the duty cycle of the instrument affects its efficiency. The duty cycle is the fraction of the time the instrument can convert the sample to ions that can be ultimately detected. The duty cycle of batch instruments that are analyzing a continuous sample stream can often be improved by a combination of continuous sample ionization and ion storage between ion batch introductions.
The utility of a mass spectrometric analysis can be significantly enhanced by performing two (or more) stages of mass analysis in tandem. A two-stage instrument is an MS/MS instrument, which performs two (or more) independent mass analyses in sequence. In the most frequently used mode of MS/MS, ions of a particular m/z value are selected in the first stage of mass analysis from among all the ions of various m/z values formed in the source. The selected ions (referred to as precursor ions) are energized, usually by collision with a neutral gas molecule, to induce ion dissociation. The ionic products of these dissociations are sorted into a product-ion mass spectrum by the second stage of mass analysis.
Tandem mass spectrometers are composed of multiple mass analyzers operating sequentially in space (Reinhold and Verentchikov 2002) or a single mass analyzer operating sequentially in time. Between the two stages of mass analysis, the ions must be subjected to some mass changing reaction such as collisional dissociation so that the succeeding mass analyzer has a different distribution of m/z values to analyze. The distribution of ions produced by the sample is called the precursor mass spectrum and is the same spectrum produced in non-tandem instruments. For each of the precursor ion entities, there will be a distribution of reaction product ions called the product ion spectrum.
Tandem mass spectrometers provide a great enhancement in detection specificity because ions appearing at a combination of precursor and product m/z values are more specific to a particular analyte than just the precursor m/z value. When the ion intensity for all combinations of the two m/z values is measured, a 3-dimensional array of data (precursor m/z vs. product m/z vs. intensity) is produced. From such a data set, mixtures of ions can be resolved without prior separation of their molecules and a great deal of structural information about individual compounds can be obtained. The development of MS/MS has had a huge impact on the analytical usefulness of mass spectrometry in all areas of its application.
A considerable amount of sample and time can be required to obtain the full MS/MS spectrum (intensities of all the product m/z's for each precursor m/z value). If the two mass analyzers are scanning devices, the ion intensity at every precursor/product m/z combination must be measured separately. This compounds the problem of sample use efficiency inherent in all scanning instruments.
If the two mass analyzers are the same device used sequentially (as with ion trap and FTMS instruments), ions having a particular m/z value in the precursor mass spectrum are isolated, and then reacted, and then the mass spectrum of the product ions is obtained (Roussis 2001). This process must be repeated for each m/z value in the precursor mass spectrum. The time required for this sequence compounds the duty cycle inefficiency of batch instruments. For full single MS spectrum generation, batch mass analyzers (time of flight mass spectrometry (TOF), ion trap mass spectrometry (ITD), Fourier transform mass spectrometry (FTMS), etc. have higher sample utilization efficiency and faster spectral generation rates than mass filter analyzers (linear quadrupoles, and sector analyzers).
ITD and FTMS are both batch techniques in that ions are taken in “batches” for analysis and all ions in a batch can be detected so that a full spectrum is generated for each batch. When used independently for MS/MS, all ions in a batch but those with the desired m/z are ejected, the selected ions undergo collisional fragmentation in the same cell, which fragmentation generates the ions seen in the product spectrum. These techniques are often called “tandem in time” since the same cell is used for precursor selection and product ion spectrum generation. The ITD uses RF voltages for ion containment within the cell and the FTMS uses a strong magnetic field. They also have different methods for ion detection.
Sample utilization efficiency is the fraction of the sample that can be converted into detectable ions. Sample utilization efficiency is adversely affected by the rejection of ions of a sample through the use of mass filters or the inattention of the instrument to the introduction of sample because it is doing something else. An example of the latter is the ITD that may be doing precursor selection and product spectrum generation while new sample is still being introduced to the ion source or sent to waste.
For many applications of mass spectrometry, desired information needs to be provided while using as small an amount of sample as possible. The range of applications and the number of days spent culturing cells and the size of an animal required for drug metabolism tests all depend on how small an amount of sample is needed to provide the desired information. This is why, for full spectrum generation, higher sample utilization efficiency and faster spectral generation rates are preferable. Regarding spectral generation rate, the preferred mode of sample introduction is through liquid chromatography, a technique in which the sample components are sorted according to their retention time on a column through which they pass. As the various compounds leave the column and flow into the source each is present for some 10's of seconds or less. This is then the amount of time available to get all the information about an eluting compound. Further, compounds often overlap in their elution. Rapid spectral generation may enable the generation of each compound's elution profile and thus allow overlapping compounds to be separately identified.
A mass filter mass analyzer (such as a quadrupole) allows transmission of ions having only a narrow range of m/z values at a time. To obtain a spectrum, there must be a steady supply of ions to the mass filter while the mass filter is scanned over the range of m/z values of interest. It is wasteful of ions relative to the “batch” analyzers (TOF, ITD, FTMS) for which all ions in a batch can be detected and assigned the appropriate m/z value.
The great success of the tandem combination of quadrupole and time-of-flight mass analyzers (an instrument called a Q-TOF) is due to the ability of the time-of-flight analyzer to produce product spectra at such a high rate that the full MS/MS spectrum can be obtained in one rather slow sweep of the quadrupole mass analyzer m/z setting. The duty cycle problem of the time-of-flight mass analyzer can be offset by introduction of an ion storage device immediately preceding it (Van Fong, 2001). Still, the poor sample utilization efficiency of the scanning quadrupole device and the relatively long time to scan through the range of desired precursor m/z values remain as limitations of this very popular instrument.
Tandem TOF instruments can reduce this problem to some extent (Barofsky 2002), though they are still only capable of generating one product spectrum for each selected precursor m/z value selected. The advantage of the TOF—TOF arrangement over the Q-TOF is principally the faster access to specific precursor m/z values and potentially faster generation of the full MS/MS spectrum.
Several researchers have conceived variations on the time-of-flight mass spectrometer in which all the precursor ions are subject to the fragmentation mechanism without preselection and the product mass is then determined by subsequent acceleration. The identification of the product ion's precursor mass is then made by the time difference in the detection of the ionic and neutral products of the fragmentation (Alderdice, Derrick et al. 1993), or by the time difference between the time of fragmentation and the time of product detection (Wollnik 1993). These approaches are very efficient in sample utilization, but have the problem that the ion flux must be maintained low in order to make the required time correlations. Such a low ion flux is inconsistent with application of the device for chromatographic detection and rapid screening of complex mixtures.
Conventional MS/MS instruments have no way to keep the information about the precursor m/z once the ion has been fragmented. Therefore, one must fragment ions of only one m/z value at a time, passing the fragments of the selected m/z value ions on to the second stage of mass analysis. Regardless of the type of mass analyzer used for the first stage of MS in an MS/MS instrument, it is therefore used as a mass filter in that only ions of only a narrow range of m/z values are accepted from it at one time. This is wasteful of sample because to obtain the product spectrum from ions that have other m/z values, one must repeat the experiment again to produce ions from each different precursor m/z value. If, while the desired set of precursor values are being selected, fragmented and analyzed, the sample composition in the source is changing (as could be the case with liquid chromatograph introduction) this adds further complication to the data analysis.
A vision of many researchers has been to obtain the full MS/MS spectrum without use of any scanning mass analyzers, producing, for each batch sample ions, the full 3-dimensional data array (McLafferty 1983, and Conzemius and Svec 1990). It would be desirable to provide a device that will do just that.
Time-of-flight mass analyzers have been previously used for product ion dispersion in MS/MS instruments. In such instruments, the first mass analyzer has sometimes been a quadrupole (Bateman and Hoyes 2000; Whitehouse and Andrien 2001), TOF, sector and other forms of mass analyzers have also been used for the selection of precursor ion m/z values. As discussed previously, ions of only a narrow range of m/z values are allowed to undergo the mass changing reaction at a time in such systems. It would be advantageous to provide a device in which each whole batch of ions would undergo fragmentation together and then be dispersed in such a way that the precursor m/z information is retained for each product ion detected.
Deconvolution is the resolving of the signals from components whose chromatographic peaks overlap into the signals each compound would have generated if it were present alone. This can be accomplished with overlapping chromatographic peaks only if the spectral information is obtained at the rate of 20 to 50 times per peak width of the eluting compounds. Until now, this has only been accomplished for 2-d (intensity vs. m/z) mass spectra. An aspect of this invention is the availability of the full 3-d spectral data on a time scale suitable for chromatographic deconvolution. The additional dimension provided by the MS/MS data should make deconvolution still more effective for complex mixture analysis. For present liquid chromatography and MS/MS, it is desirable to obtain the full 3-d MS/MS information several times every second or even more often. As improvements in chromatography shorten the peak widths, rapid spectral generation will become even more important. Another aspect of the invention with respect to chromatographic deconvolution is that all the MS/MS data are collected for the same batch of ions from the source so that there will be no difference in chromatographic time among elements of the data used in the deconvolution step. This lack of spectral skew is very valuable in the application of deconvolution algorithms.